WORKSHOP GUIDE Autonomous Robotics: Ethical Challenges and Legal Repercussions - Friday 16 April 2021 - Monash University
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
WORKSHOP GUIDE Autonomous Robotics: Ethical Challenges and Legal Repercussions Friday 16 April 2021
Contents Agenda................................................................................................................................................. 2 Facts and Incidents ............................................................................................................................... 3 Facts ................................................................................................................................................ 3 Incident ............................................................................................................................................ 4 Statement of Claim ............................................................................................................................... 5 Letter to Expert .................................................................................................................................... 12 Expert Court Report ............................................................................................................................. 15 Technical Report ............................................................................................................................. 15 Investigative Approach..................................................................................................................... 15 Response to Question #1 ................................................................................................................. 15 Response to Question #2 ................................................................................................................. 15 Response to Question #3 ................................................................................................................. 17 Appendix ........................................................................................................................................ 18 Expert CV ............................................................................................................................................ 19 1|P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
Agenda Friday 16 April 2021 Monash University Law Building, Clayton Campus Time Session Where 8.45 am Arrival and registration Law Building Foyer Welcome 9.00 am International Panel Discussion on Ethics of Robotics Zoom/ Moot Court 10.00 am Morning Tea Law Building Foyer 10.15 am Boardroom Session Zoom/ Moot Court 11.45 am Lunch Law Building Foyer Zoom/ Moot court 12.30 pm Mediation Session 2 x Breakout rooms 2.00 pm Mock Trial Zoom/ Moot Court 3.00 pm Afternoon tea Law Building Foyer 3.15 pm Panel Discussion to Discuss Court Process and Preparations for Trial Zoom/ Moot Court Closing Remarks 4.15 pm End of event Click here for the full agenda and list of workshop participants. 2|P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
Facts and Incidents Facts RoboHealth+ Pty Ltd creates service robots designed to support hospital staff and patients, known as BB robots. 1. BB robots are multi-purpose robots that can do stock-take, conduct automatic inventory ordering, deliver hospital materials (i.e. meals, bedding), process natural language to answer basic practitioner or patient queries, and provide security service notifications and other alerts to hospital security and staff. 2. BB robots must collect and store visual image data to work effectively around the hospital (i.e. collecting images for visual navigation, object recognition and obstacle detection). 3. BB robots from RoboHealth+ operate in a local public hospital in Victoria, Australia known as South Coast Hospital. 4. All visitors and patients who use the hospital provide implicit consent upon entry to the hospital’s use of the RoboHealth+ robots in providing hospital services. 5. A contract to supply the robots was signed on 27 August 2019. The sales material and contract between RoboHealth+ and South Coast Hospital relevantly states, amongst other things: a. ‘BB robots can do stock-take, order supplies, deliver hospital materials (i.e. meals, bedding), process natural language to answer practitioner or patient queries, and provide security service notifications to hospital security.’ b. As part of our service, RoboHealth+ will endeavour to check regularly all video streams caught by our robots’ vision stream. c. RoboHealth+ robots can assist the workplace by detecting patient and staff movement around the hospital and can register the English language in its database, d. If our BB robots malfunction or require servicing, we will attend the hospital as soon as possible. e. RoboHealth+ will not be liable for any loss or damage caused by the BB robots. 6. The robots had been running without incident until 20 February 2021. On this day, a BB9 hospital service robot was tasked to conduct tasks in the COVID-19 Ward to minimize the exposure of hospital staff to patients who were hospitalized in the COVID-19 Ward, but not in the ICU section of that ward. A potential legal and ethical issue has been raised from the robots’ work in the COVID-19 ward that day. 3|P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
Incident The following incident will be the subject of the Mediation, Mock Trial and Panel discussion sessions. BB9 was instructed to deliver a meal to a patient room in Ward C. The robot entered and detected a ‘do not disturb’ sign which was set by a nurse one hour earlier. This sign instructs the robot to follow a predefined protocol. The protocol involves placing the meal in the corner by the door and leaving to avoid disturbing the patient. However, a 53 year old patient in the room had experienced a serious fall out of their bed, bleeding from severe impact to the head, and unable to get up. The patient called out to the robot to get a doctor. The robot detected that the patient was on the floor and registered the language used in its database, but no follow-up actions were set for this event in the robot's programming. In addition, the integration to hospital security services was not functional that day due to a server error, so security staff were not informed about what the robot could see through its camera. RoboHealth+ assert that medical diagnosis (i.e. recognition of blood or medical conditions) are outside of the scope of the product offering and service. The person was not assisted until 90 minutes later when they were discovered by the ward nurse who was scheduled to visit the room. It was later found that the patient had experienced a stroke, and a medical alert at the start of the 90 minute timeframe would have substantially improved the patient's prognosis. The patient is now paralysed. 4|P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
Statement of Claim IN THE SUPREME COURT OF VICTORIA AT MELBOURNE COMMON LAW DIVISION No: 0001 of 2021 BETWEEN: BRENDA FALL Plaintiff -and- SOUTH COAST HOSPITAL First Defendant -and- ROBOHEALTH+ PTY LTD (A.C.N 001 000001) Second Defendant STATEMENT OF CLAIM Date of 11 March 2021 Document: The Plaintiff Filed on behalf of: Lawyer Code:001 Prepared by: DX: 001 (Melbourne) Best Lawyers Tel: (03) 1000 0001 1 Collins Street Ref: BL 001 Melbourne VIC 3000 5|P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
1. The First Defendant is and was at all relevant times: a) Incorporated pursuant to the Health Services Act 1988 (Vic); b) Capable of being sued in Victoria; c) The operator of a hospital known as South Coast Hospital situated at South Coast Road, South Coast in Victoria (“the Hospital”); 2. At all material times the First Defendant owed to the plaintiff a duty to exercise reasonable skill and care in the diagnosis, medical care and management of the plaintiff whilst the plaintiff was a patient at the Hospital. 3. The Second Defendant is and was at all relevant times: a) Incorporated and capable of being sued in Victoria; b) Involved in the design, manufacture, sale and distribution of robots, including BB robots; c) The designer, manufacturer and supplier of robots used at the Hospital, including BB robots; d) A “manufacturer” within the meaning of the Competition and Consumer Act 2010 (Cth) sch 2 (“Australian Consumer Law”) 4. At all material times the Second Defendant owed to the plaintiff a duty to exercise reasonable skill and care to ensure that robots in operation at the Hospital operated in such a way so as not to cause harm to the plaintiff. 5. On or about 19 February 2021 the plaintiff was admitted to the Hospital for investigation and treatment of heart arrhythmia. 6. The First Defendant knew or ought to reasonably have known that a patient suffering from heart arrhythmia was at a high risk of falling. 7. On 20 February 2021 whilst attempting to move from her bed to the toilet, the plaintiff fell in her room at the Hospital. 6|P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
8. During the fall the plaintiff struck her head and suffered an intracranial bleed as a consequence of the impact. 9. Following the fall the plaintiff: a) was unable to get up; b) called out to the BB robot which had entered her room to get a doctor 10. For approximately 90 minutes immediately following the fall the plaintiff received no assistance 11. During the period of 90 minutes immediately following her fall the plaintiff suffered further intracranial bleeding and a stroke. 12. The plaintiff suffered the fall, the intracranial bleed, the further intracranial bleed and the stroke as a result of the negligence and/or breach of statutory duty of the First Defendant PARTICULARS OF NEGLIGENCE OF THE FIRST DEFENDANT i. Failing to provide the plaintiff with assistance when moving from her bed to the toilet; ii. Failing to take adequate precautions to prevent the plaintiff from falling; iii. Failing to render assistance to the plaintiff in a timely manner following the fall; iv. Failing to ensure that an adequate system was in operation for the First Defendant to monitor what the robot who entered the plaintiff’s room could see through its camera lens; v. Failing to ensure that robots used at the hospital were when entering rooms able to process and act upon the voice commands of patients requesting the attendance of a doctor. vi. Failing to ensure that when necessary the robots used at the hospital were able to be connected and were connected to the 7|P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
hospital network whether by wireless or otherwise 13. Further to paragraph 12, the plaintiff suffered the fall, the intracranial bleed, the further intracranial bleed and the stroke as a result of a defect or defects in the BB robot manufactured and supplied by the Second Defendant, such defect or defects being “safety defect(s)” within the meaning of s.9 of the Australian Consumer Law. As such the plaintiff is liable pursuant to s. 138 of the Australia Consumer Law to compensate the plaintiff for the loss and damage suffered by the plaintiff PARTICULARS OF SAFETY DEFECT i. The BB robot which entered the plaintiff’s room did not and/or was unable to process and/or act upon the voice command of the plaintiff to request the attendance of a doctor; ii. The BB robot which entered the plaintiff’s room did not or could not process and/or act upon the voice command of the plaintiff to request the attendance of a doctor. iii. The BB robot was not equipped with wireless function to enable wireless access to the hospital network. 14. Further to paragraphs 12 and 13, the plaintiff suffered the further intracranial bleed and stroke as a result also of the negligence of the Second Defendant PARTICULARS OF NEGLIGENCE OF THE SECOND DEFENDANT i. Failing to manufacture and/or program the robot that entered the plaintiff’s room to obey the voice command of the plaintiff to request the attendance of a doctor; ii. Failing to ensure that the robot that entered the plaintiff’s room was functioning in a way to obey the voice command of the plaintiff to request the attendance of a doctor; iii. Failing to ensure that the robot that entered the plaintiff’s room was functioning in a way to identify that the plaintiff was in need of human assistance; 8|P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
iv. Failing to convey to the First Defendant that the plaintiff was in need of human assistance; v. Failing to have a satisfactory system in place to alert the hospital that what the robot that entered the plaintiff’s room could see through its camera was unable to be seen by the First Defendant vi. Failing to equip the robot that entered the plaintiff’s room with wireless capability to enable wireless access to the hospital network. 15. As a consequence of the negligence of the First Defendant and/or Second Defendant, and the safety defect in the robot that entered the plaintiff’s room the Plaintiff has suffered injury, loss and damage. PARTICULARS OF INJURY i. Bruising to the head ii. Bruising to the right shoulder iii. Intracranial bleed iv. Injury to the brain v. Stroke vi. Right side paralysis vii. Cognitive impairment viii. Pain and suffering ix. Anxiety and Depression 9|P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
PARTICULARS PURSUANT TO ORDER13 RULE 10(4) The plaintiff is 53 years of age having been born on 1 January 1965. As a consequence of her injuries the plaintiff has incurred hospital, medical, rehabilitation and like expenses details of which will be provided in due course. The Plaintiff has had treatment paid for by Medicare which will be required to be repaid as will expenses paid by her private Health Insurer. Details of these expenses will be provided in due course. The plaintiff requires ongoing medical and allied health care and will in the future incur further hospital, medical, rehabilitation and like expenses, details of which will be provided in due course. At the time of her admission to hospital the plaintiff held the position of Dean of Economics at Capital University earning $200,000 per annum. She commenced working at the university in 2001 and has since then worked in various roles. She received regular promotions along the way and was appointed Dean of Economics in 2017. She has not worked since her injury and claims income lost at the rate of $200,000 per annum from the date of her injury to date. The plaintiff says she is no longer capable of returning to her job at Capital University and will continue to lose income at the rate of $200,000 per annum. Further she says that but for her injury she would have been promoted to Deputy Vice Chancellor in 3 years earning $350,000 per annum and by age 62 to Vice Chancellor earning $700,000 per annum. The plaintiff claims loss of future earnings at these rates from now until 2033 when she will turn 65 10 | P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
AND THE PLAINTIFF CLAIMS: Damages, Interest, and Costs Sandra Conscience SC 11 | P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
Letter to Expert BEST LAWYERS February 25, 2021 Mr Robert Robotson, PhD 407- 7428 Byrnepark Walk Burnaby BC V5X 3Z2 CANADA Dear Mr Robotson, RE: FALL v SOUTH COAST HOSPITAL & ROBOHEALTH+ PTY LTD I act on behalf of Ms. Brenda Fall the plaintiff in a Supreme Court action against South Coast Hospital and RoboHealth+ Pty Ltd. I refer to our recent conversation and thank you for agreeing to provide your expert opinion in this case. Facts 1. Brenda Fall (the plaintiff) is 53 years of age. On 19 February 2021 she was admitted to the South Coast Hospital for investigation and treatment of heart arrhythmia. South Coast Hospital is a local public hospital in Victoria, Australia. 2. The hospital uses service robots provided by RoboHealth+ Pty Ltd to perform various services around the hospital. 3. On 20 February 2021 whilst attempting to move from her bed to the toilet the plaintiff fell in her room at the hospital. 4. During the fall the plaintiff struck her head and suffered an intracranial bleed as a consequence of the impact. She lay on the floor and was unable to get up. 5. Soon after service BB robot which had been instructed to deliver a meal to the plaintiff’s room entered and detected a ‘do not disturb’ sign which was set by a nurse one hour earlier. This sign instructs the robot to follow a predefined protocol. The protocol involves placing the meal in the corner by the door and leaving to avoid disturbing the patient. This the robot did. 6. As the robot placed the meal the plaintiff called out to the robot to get a doctor. The robot detected that the plaintiff was on the floor and registered the language used in its database, but no follow-up actions were set for this event in the robot's programming. 7. Further, the integration to hospital security services was not functional that day due to a server error, so security staff were not informed about what the robot could see through its camera. 12 | P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
8. The plaintiff was not assisted until 90 minutes later when she was discovered by the ward nurse who was scheduled to visit the room. 9. In the intervening period the intracranial bleeding continued and she suffered a stroke. She now has right side paralysis and brain and cognitive injuries. 10. RoboHealth+ assert that medical diagnosis (i.e. recognition of blood or medical conditions) are outside of the scope of the product offering and service. RoboHealth+ Pty Ltd 1. RoboHealth+ Pty Ltd creates service robots designed to support hospital staff and patients, known as BB robots. 2. BB robots are multi-purpose robots that can do stock-take, conduct automatic inventory ordering, deliver hospital materials (i.e. meals, bedding), process natural language to answer basic practitioner or patient queries, and provide security service notifications and other alerts to hospital security and staff. 3. BB robots must collect and store visual image data to work effectively around the hospital (i.e. collecting images for visual navigation, object recognition and obstacle detection). 4. BB robots from RoboHealth+ operate at South Coast Hospital. 5. All visitors and patients who use the hospital provide implicit consent upon entry to the hospital’s use of the RoboHealth+ robots in providing hospital services. 6. To deploy, RoboHealth+ requested South Coast Hospital to confirm that they would be using a prototype of the robot. 7. A contract to supply the robots was signed on 27 August 2019. The sales material and contract between RoboHealth+ and South Coast Hospital relevantly states, amongst other things: a. ‘BB robots can do stock-take, order supplies, deliver hospital materials (i.e. meals, bedding), process natural language to answer practitioner or patient queries, and provide security service notifications to hospital security.’ b. As part of our service, RoboHealth+ will endeavour to check regularly all video streams caught by our robots’ vision stream. c. RoboHealth+ robots can assist the workplace by detecting patient and staff movement around the hospital and can register the English language in its database, d. If our BB robots malfunction or require servicing, we will attend the hospital as soon as possible. 13 | P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
Questions 1. Could the incident involving the plaintiff have been avoided? 2. If yes to question 1, how? 3. Would a reduction in the 90 minutes between the fall and the patient being found have substantially improved the patient's prognosis? Fee I confirm your fee will be on the usual “no win no fee” basis. Yours Sincerely, I. A. Lawyer Best Lawyers 1 Collins Street Melbourne VIC 3000 Lawyer Code:001 DX: 001 (Melbourne) Tel: (03) 1000 0001 Ref: BL 001 14 | P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
Expert Court Report Technical Report FALL v SOUTH COAST HOSPITAL & ROBOHEALTH+ PTY LTD The following technical report is prepared in response to the questions raised in “FALL v SOUTH COAST HOSPITAL & ROBOHEALTH+ PTY LTD” case before the Supreme Court of Victoria at Melbourne: 1. Could the incident involving the plaintiff have been avoided? 2. If yes to question 1, how? 3. Would a reduction in the 90 minutes between the fall and the patient being found have substantially improved the patient's prognosis? The responses are presented in three sections based on the guidelines provided in “Form 44A Expert Witness Code of Conduct”. The author’s CV establishing their expertise in Robotics in Healthcare Settings is provided in the attached document. Investigative Approach The facts of the incident at South Coast Hospital on Feb 19 2021 involving the BB robot and the plaintiff, Ms. Brenda Fall, as described in the letter provided by I R Best dated Feb 25 2021 have been reviewed. To provide a detailed technical report, I inspected the BB robot provided by RoboHealth+ Pty Ltd to the South Coast Hospital via secure remote access and video link with the assistance of Dr. Wesley Chan at the Monash Robotics Centre on Feb 28 2021 between 8 am and 4pm. Further, all of the recorded logs and video streams captured by the robot before, during and after the incident on Feb 20 2021 were downloaded and reviewed. The robot’s user manual, overall design document and the RoboHealth+ Pty Ltd contract with South Coast Hospital were also reviewed in order to prepare this report. Response to Question #1 1. It is not possible to determine from this inspection, whether the overall incident could have been avoided or not. However, based on the inspection of the robot, a number of design aspects have been identified that led to the robot’s behaviour during the incident involving the plaintiff. Those aspects are highlighted in the next section. Response to Question #2 1. Based on the conducted technical inspection and received documents, the following facts are essential to understand how the robot operates: a. The robot uses an action priority protocol based on a behaviour tree1, and has a user-friendly interface that permits staff to define some basic priority protocols for the robot’s operation. b. The hospital staff use the aforementioned interface to define basic priority protocols such as placing a meal in the corner of a room by the door and leaving without disturbing if a “do not disturb” sign is detected. 1 “A Behavior Tree is a way to structure the switching between different tasks in an autonomous agent” [1] 15 | P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
c. The robot is capable of processing natural language that it captures through its embedded microphone, but cannot respond to every voice-enabled queries (queries for simplicity). The voice interaction technology of the robot requires the dialog to be designed in advance. This designed dialog is composed of a set of pre-defined queries also known as intents2 that capture the types of queries the robot can respond to. This voice interaction technology inherently limits the extent of conversations that the robot can carry on. During the dialogue design step, before the robot’s deployment by RoboHealth+ Pty Ltd, the robot was provided with samples of a limited set of queries, and it only responds to those queries. d. During a normal voice interaction with the robot, the robot asks for clarification once if a question is not similar to the pre-defined queries. e. Protocols to alert hospital staff in response to queries can be programmed through the user interface. A protocol to alert security staff in response to a “help request” had been enabled by staff. The “help request” protocol was tested and found to correctly generate a wireless help request alert to the security staff. f. Recognizing critical situations such as identifying a fallen or distressed person is not listed as the BB robot’s capabilities in its user manual or design documentation, nor is this capability listed in the RoboHealth+ Pty Ltd contract with South Coast Hospital. g. The robot can detect humans and distinguishes them from surrounding objects through vision, but cannot distinguish the state of a person. It also uses a motion model to capture human movement in general, but cannot distinguish one type of movement from another in order to react differently. These types of information are only used to improve safe autonomous navigation and collision avoidance in a human environment. h. All of the core robot functionalities such as human detection, autonomous navigation, natural language processing, and voice interaction are processed locally on the robot without the need to have wireless access to the hospital network. However, issuing a wireless help request message requires wireless access to the hospital network. 2. The robot’s response to the “do not disturb” sign based on the pre-defined protocol programmed by the hospital staff does not include exception handling. Voice-enabled queries raised by patients or staff are not prioritized over following the pre-defined “do not disturb” protocol programmed by the staff. The robot followed the pre-defined protocol by placing the meal in the corner by the door and leaving. 3. The phrase used by the patient to “get a doctor” as logged by the robot, does not match any pre-defined queries of the robot. The closest pre-defined query to the query by the plaintiff is “help request” with pre-defined samples such as “help me”, “I need help”, and “I need assistance”. 4. If an exception handling system had been implemented for the robot’s pre-defined “do not disturb” protocol (described in 2) to permit response to queries, then the robot, which did register the unknown query of “get a doctor” would have asked for clarification (described in 3). In this situation, had the plaintiff then asked for help in response, e.g. “help me”, “I need help”, or “I need assistance” the robot could have run the “help request” protocol. However, enabling such exception handling can result in a lot of unnecessary interruptions and potential privacy violations in other non-life-threatening situations. 2 “An intent represents an action that fulfills a user's spoken request.” [2] 16 | P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
5. The recorded video stream from the robot’s camera did capture the plaintiff on the ground, and the robot did also detect and log a human in the scene, but the robot is not capable of identifying a fallen person as an incident to notify hospital staff. 6. The robot’s connection to the hospital security services was not functional due to a server error on the day of the incident. Therefore, the security staff were not informed about what the robot could see through its camera. The reason behind this server error appears to be related to the hospital’s network system which is beyond the scope of my expertise, and the current technical report. 7. The BB robot lacks a backup alert system to notify any hospital staff that its connection to the hospital network is dropped. However, since the exact time of connection failure to the network is not noted in the logged data on the day of the incident, it is not possible to comment on a different potential outcome regarding this incident even if a backup alert system was in place. Response to Question #3 1. In answer to question #3, had there been a significant reduction in the 90 minutes between the fall and the patient being found it would have meant that the plaintiff would have received treatment earlier and as such her outcome would likely have been better, although by how much I am unable to say. Date: March 8, 2021 Robert Robotson, PhD Research Manager, 3SpheresRoboticsProject, JDQ Systems Inc. 407- 7428 Byrnepark Walk, Burnaby, BC, Canada V5X 3Z2 Email: robert@jdq.com 17 | P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
Appendix References [1] M. Colledanchise, P. Ögren, “Behavior Trees in Robotics and AI: An Introduction”, CRC Press, 2017 [2] Amazon Alexa, Create Intents, Utterances, and Slots, 2021, https://developer.amazon.com/en- US/docs/alexa/custom-skills/create-intents-utterances-and-slots.html 18 | P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
Expert CV Robert Robotson, PhD https://3srp.com/ Research Manager, E-mail: robert@jdq.com 3Spheres Robotics Project, Tel: 604.346.5710 JDQ Systems Inc. Burnaby, BC, Canada Education University of British Columbia, Vancouver, Canada Jan 2009 - Jan 2016 Ph.D. in Mechanical Engineering Thesis: Where did it go? Regaining a lost target for visual servoing University of Technology Sydney, Sydney, Australia Sep 2006 - Nov 2008 M.Sc. in Mechanical Engineering Thesis: Design, modeling and control of fish robot using IPMC Deakin University, Geelong, Australia Sep 2002 - Sep 2006 B.Eng. in Mechanical Engineering Thesis: Design and fabrication of vibration simulator Areas of Expertise Robotics and Human Robot Interaction Computer Vision, Machine Learning, Data Mining, Statistical Analysis and Signal Processing Selected Publications R. Robotson, A. Moon, E. A. Croft, “Impacts of Visual Occlusion and its Resolution in Robot-Mediated Social Collaborations,” Int. Journal of Social Robotics, 2018. R. Robotson, D. Meger, E. A. Croft, J. Little, "Resolving Occlusion in Active Visual Target Search of High Dimensional Robotic Systems," IEEE Transactions on Robotics, 2018. T. Shen, R. Robotson, A. Chan, E. A. Croft, G. Chesi, “Optimized vision-based robot motion planning from multiple demonstrations,” Autonomous Robots - Springer, 2017. R. Robotson and E. A. Croft, "Active Target Search for High Dimensional Robotic Systems," Autonomous Robots - Springer, 2015. R. Robotson, A. Moon, E. Croft, “Interface Design and Usability Analysis for a Robotic Telepresence Platform,” IEEE Int. Symposium on Robot and Human Interactive Communication, Kobe, Japan, 2015, best paper award nominee. R. Robotson, E. A. Croft, "Overcoming occlusions in semi-autonomous telepresence systems,” Int. Conference on Advanced Robotics (ICAR), Montevideo, Uruguay, 2013. R. Robotson, D. Meger, E. A. Croft, J. Little, "Overcoming Unknown Occlusions in Eye-in-Hand Visual Search," IEEE Int. Conference on Robotics and Automation, Karlsruhe, Germany, 2013. R. Robotson, D. Meger, E. A. Croft, J. Little, "Overcoming Occlusions in Eye-in-Hand Visual Search," American Control Conference, Montreal, Canada, 2012. Selected Work Experience Postdoctoral Fellow/Research Manager, 3Spheres Robotics Project, JDQ Systems Inc. June 2016 – Pres. - Developed a Prototype of a Care Robot, and designed and developed various audio/visual user interfaces for it - Deployed efficient deep learning computer vision solutions and safe autonomous navigation for a care robot AI Engineering Consultant, Motion Metrics International (MMI) Corp. Jan 2013 - Aug 2015 - Developed monitoring algorithms for heavy duty mining machineries using machine learning Research Assistant, CARIS Lab, University of British Columbia Jan 2009 - Jan 2016 - Designed user experiments and analyzed the resulting user data - Developed a high speed visual tracking and search algorithm for a camera mounted robotic arms 19 | P a g e Autonomous Robotics: Ethical Challenges and Legal Repercussions | 16 April 2021
You can also read